Anomaly Detection in WBAN Research Topics

Anomaly Detection (AD) in WBAN research topic is now widely utilized to detect malicious or unauthorized attacks. Various technologies and parameters were utilized in this research to get a robust outcome. Below we provide some WBAN based anomaly detection related concepts, applications, techniques and parameters.

1. Define Anomaly Detection in WBAN

At the first stage we look over the definition for WBAN based AD. WBANs are unique networks that are created for monitoring healthcare and they generally consist of sensors which measure different physical parameters like body temperature, motion, heart rate and other vital signs. AD in WBAN describes the procedure to find irregular or uncommon events, patterns or behaviors over the data gathered from sensors surrounded in wearable devices or attached to the human body.

2. What is Anomaly Detection in WBAN?

Subsequent to the definition we see the in-depth description for this proposed research. AD in WBAN describes the procedure to find abnormalities or uncommon patterns in the data gathered from sensors over a WBAN setup. To find the irregular patterns or events Anomaly detection in WBAN is important for healthcare applications which will specify the possible health emergencies or problems.

3. Where Anomaly Detection in WBAN used?

Afterwards the in-depth description we discuss where to use this WBAN technology. It will have applications in industrial settings for worker monitoring and protection. In smart environments, WBANs are combined into smart clothing or surrounded in the environments to allow the background-aware computing and to improve the efficacy of different procedures.

4. Why Anomaly Detection in WBAN technology proposed? , previous technology issues

We propose Anomaly detection in WBAN technology and it resolves many of the difficulties in the healthcare monitoring system. The following are the issues that it will handle:

• Both reliability and adjustability of anomaly detection will be dangerous in centralized RSU (Roadside Unit) systems by issues like single points of loss.
• The sensitivity of the device to changes in the physical state of specific and the accuracy of the ECG measurement that will result in false negatives or positives in anomaly detection.
• The latency among networks, shortages of resources, confidentiality of data and the sensitivity of physical indicators are some of the frequent problems around healthcare monitoring.
• Actual time usability and time are negatively affected by the computational resource intensity of implementing the deep learning models, ECG reading analysis and filtering algorithms.
• Problems on the basis of probable loss diagnostic accuracy emerge when contrasted to the standard 12-lead ECGs, because of the minimization of leads in the regard of patient suitability.
• It is very challenging to evaluate the image encryption technique’s reality and locate probable issues because of the absence of details that underlines the requirement for detailed evaluation.
• The flexibility of anomaly detection technique is more hindered by control evaluating and dependence on the device accuracy.
• The issues in identifying and categorizing the anomalies in WBAN for healthcare monitoring are very challenged by cost limits, susceptibility in untrusted networks and limited network coverage. These factors highlight the requirement to solve the scaling problems.
• The extreme complexity nature of the physical data and resource restrictions will supplement the additional layers of complexity which affects the perfect approach to construct the monitoring model.

5. Algorithms / protocols

In this research the WBAN based AD technique is proposed and it addresses some existing technology issues. Now we give several methods that are used in this research are Anomaly Detection using Deep Learning approach, Anomaly Detection using Statistical approach and Anomaly detection using Machine Learning Approach.

6. Comparative study / Analysis

We propose a WBAN based Anomaly Detection in this research and are compared with the various methods to obtain the best findings. The methods that we compared are Anomaly Detection using Machine Learning approach, Deep Learning approach and Statistical Approach.

7. Simulation results / Parameters

Some of the previous works are simulated by utilizing the language python.

8. Dataset LINKS / Important URL

WBAN based AD technology is proposed in this research and it addresses some existing technology issues by utilizing the following links. Here we offer links that are related to our proposed research and we can overview the links to clarify the doubts.

• https://ieeexplore.ieee.org/abstract/document/9204697/
• https://dl.acm.org/doi/abs/10.1145/3501773
• https://link.springer.com/article/10.1007/s12652-018-1091-x
• https://link.springer.com/article/10.1007/s12652-019-01343-w
• https://ieeexplore.ieee.org/abstract/document/9954371/

9. Anomaly Detection in WBAN Applications

There are many applications to be used for this proposed WBAN based AD. Some of the applications are medicine, Home-based rehabilitation health, Remote health monitoring, Sensor diversification, Home/health care, Real-time feedback, Sports and Multimedia.

10. Topology for Anomaly Detection in WBAN

Now we see the topology for this proposed technology to monitor healthcare. The topology for categorizing and anomaly detection comprises a hierarchical framework. Physical sensors like accelerometers and heart rate monitors form the sensor layer, producing the data which go through the feature extraction and signal processing. Then the multimodal fusion combines data along different sensors. Then Anomaly detection, executed through the machine learning models and the thresholding techniques, finds the abnormalities from normal patterns. The categorization module, frequently based on the supervised learning models, classifies particular anomalies. Decision-making elements will evaluate the risk of anomalies and then it generates alarms. Edge or Cloud computing will be utilized for processing and preservation, with encryption to make sure of the data protection.

11. Environment for Anomaly Detection in WBAN

Anomaly identification and categorization in WBAN for healthcare monitoring systems will generate in an active and complex environment. Over this framework, the wearable or adaptable sensors will frequently gather different physical data from everyone, which comprises temperature, motion patterns and heart rate. The environment needs the actual-time processing capacities to correctly find the abnormalities from the recognized standards that signify possible health problems.

12. Simulation tools

Here we see the software requirements that are employed for this research. The tool that is required for this research is Python 3.11.4 or above version, this tool is used to implement our proposed research. Then the research is operated by employing the operating system namely Windows – 10 (64- bit).

13. Results

The WBAN based AD is proposed in this research and it overcomes several existing technology issues. In this we detect any anomalies or unauthorized patterns or any other data that were collected from the sensor. This technique can be implemented by employing tools like Python 3.11.4 or above version.

Anomaly Detection in WBAN Research Ideas:

The following are the research topics that are related to our proposed technique anomaly detection in WBAN. This topic gives some relevant information to clarify the details about this proposed research.

1. Anomaly Detection in Wireless Body Area Network using Mahalanobis Distance and Sequential Minimal Optimization Regression
2. Markov Models for Anomaly Detection in Wireless Body Area Networks for Secure Health Monitoring
3. Isolation Forest Based Anomaly Detection Approach for Wireless Body Area Networks
4. Anomaly Detection in WBANs Using CNN-Autoencoders and LSTMs
5. A Correlation-Based Anomaly Detection Model for Wireless Body Area Networks Using Convolutional Long Short-Term Memory Neural Network
6. Unsupervised Anomaly Detection with Self-Training and Knowledge Distillation
7. Performance Comparison of Soiling Detection Using Anomaly Detection Methodology
8. Target-to-Anomaly Conversion for Hyperspectral Anomaly Detection
9. Hyperspectral Anomaly Detection with Data Sphering and Unsupervised Target Detection
10. Error Distribution-based Anomaly Score for Forecasting-based Anomaly Detection of PV Systems
11. Hyperspectral Anomaly Detection for Spectral Anomaly Targets via Spatial and Spectral Constraints
12. Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progress (Extended Abstract)
13. Hyperspectral Anomaly Detection via Background Purification and Spatial Difference Enhancement
14. Enhancing Hyperspectral Anomaly Detection by Difference-of-Convex Sparse Anomaly Modeling
15. Discriminative Learning for Supervised Anomaly Detection
16. Unsupervised and Ensemble-based Anomaly Detection Method for Network Security
17. AttentionAE: Autoencoder for Anomaly Detection in Attributed Networks
18. Deep Unrolling Network with Active Dictionary Learning for Hyperspectral Anomaly Detection
19. Autoencoder in Autoencoder Network Based on Low-Rank Embedding for Anomaly Detection in Hyperspectral Images
20. Research on Time Series Anomaly Detection Based on Graph Neural Network
21. LogST: Log Semi-supervised Anomaly Detection Based on Sentence-BERT
22. Review of electric vehicles’ charging data anomaly detection based on deep learning
23. Separating Sensor Anomalies From Process Anomalies in Data-Driven Anomaly Detection
24. A Hybrid Method for Anomaly Detection Using Distance Deviation and Firefly Algorithm
25. SMD Anomaly Detection: A Self-Supervised Texture–Structure Anomaly Detection Framework
26. A Comprehensive Survey on Graph Anomaly Detection With Deep Learning
27. Spatial-Spectral Extraction for Hyperspectral Anomaly Detection
28. Enhancing Network Anomaly Detection with Optimized One-Class SVM (OCSVM)
29. OHODIN – Online Anomaly Detection for Data Streams
30. GAN-based Hyperspectral Anomaly Detection
31. Cross-Epoch Learning for Weakly Supervised Anomaly Detection in Surveillance Videos
32. Hyperspectral Anomaly Detection via Local Gradient Guidance
33. Entropy Change Rate for Traffic Anomaly Detection
34. Assessment of Real-World Incident Detection Through a Component-Based Online Log Anomaly Detection Pipeline Framework
35. A Novel Semi-supervised Anomaly Detection Method for Network Intrusion Detection
36. Anomaly Detection of Metro Station Tracks Based on Sequential Updatable Anomaly Detection Framework
37. Local Attention base on Time Masks for Weakly Supervised Video Anomaly Detection
38. TCN-Log2Vec: A Comprehensive Log Anomaly Detection Framework based on Optimized Log Parsing and Temporal Convolutional Network
39. Research on Anomaly Detection Algorithm of Chemical Storage Tanks Based on FCM-LSTM
40. Robustness Evaluation Method for Spacecraft Anomaly Detection Models
41. Anomaly Detection with Partially Observed Anomaly Types
42. Anomaly Detection of Wireless Relays Based on Markov Models Through the Wald-Wolfowitz Runs Test
43. Deep-RX for Hyperspectral Anomaly Detection
44. Lesion2void: Unsupervised Anomaly Detection in Fundus Images
45. Graph and Sparsity Regularized Decomposition Model for Hyperspectral Anomaly Detection
46. Characterization of Background-Anomaly Separability With Generative Adversarial Network for Hyperspectral Anomaly Detection
47. Subfeature Ensemble-Based Hyperspectral Anomaly Detection Algorithm
48. Real-World Video Anomaly Detection by Extracting Salient Features
49. Anomaly Detection via Context and Local Feature Matching
50. An Anomaly Detection Algorithm for Multi-dimensional Segmentation Plane Isolation Forest